def commit_nav_toc(container, toc, lang=None):
from calibre.ebooks.oeb.polish.pretty import pretty_xml_tree
tocname = find_existing_nav_toc(container)
if tocname is None:
item = container.generate_item('nav.xhtml', id_prefix='nav')
item.set('properties', 'nav')
tocname = container.href_to_name(item.get('href'), base=container.opf_name)
try:
root = container.parsed(tocname)
except KeyError:
root = container.parse_xhtml(P('templates/new_nav.html', data=True).decode('utf-8'))
et = '{%s}type' % EPUB_NS
navs = [n for n in root.iterdescendants(XHTML('nav')) if n.get(et) == 'toc']
for x in navs[1:]:
extract(x)
if navs:
nav = navs[0]
tail = nav.tail
attrib = dict(nav.attrib)
nav.clear()
nav.attrib.update(attrib)
nav.tail = tail
else:
nav = root.makeelement(XHTML('nav'))
first_child(root, XHTML('body')).append(nav)
nav.set('{%s}type' % EPUB_NS, 'toc')
if toc.toc_title:
nav.append(nav.makeelement(XHTML('h1')))
nav[-1].text = toc.toc_title
rnode = nav.makeelement(XHTML('ol'))
nav.append(rnode)
to_href = partial(container.name_to_href, base=tocname)
spat = re.compile(r'\s+')
def process_node(xml_parent, toc_parent):
for child in toc_parent:
li = xml_parent.makeelement(XHTML('li'))
xml_parent.append(li)
title = child.title or ''
title = spat.sub(' ', title).strip()
a = li.makeelement(XHTML('a' if child.dest else 'span'))
a.text = title
li.append(a)
if child.dest:
href = to_href(child.dest)
if child.frag:
href += '#'+child.frag
a.set('href', href)
if len(child):
ol = li.makeelement(XHTML('ol'))
li.append(ol)
process_node(ol, child)
process_node(rnode, toc)
pretty_xml_tree(rnode)
for li in rnode.iterdescendants(XHTML('li')):
if len(li) == 1:
li.text = None
li[0].tail = None
container.replace(tocname, root)
def beautify_text(raw, syntax):
from lxml import etree
from calibre.ebooks.oeb.polish.parsing import parse
from calibre.ebooks.oeb.polish.pretty import pretty_xml_tree, pretty_html_tree
from calibre.ebooks.chardet import strip_encoding_declarations
if syntax == 'xml':
root = etree.fromstring(strip_encoding_declarations(raw))
pretty_xml_tree(root)
elif syntax == 'css':
import logging
from calibre.ebooks.oeb.base import serialize, _css_logger
from calibre.ebooks.oeb.polish.utils import setup_css_parser_serialization
from css_parser import CSSParser, log
setup_css_parser_serialization(tprefs['editor_tab_stop_width'])
log.setLevel(logging.WARN)
log.raiseExceptions = False
parser = CSSParser(loglevel=logging.WARNING,
# We dont care about @import rules
fetcher=lambda x: (None, None), log=_css_logger)
data = parser.parseString(raw, href='<string>', validate=False)
return serialize(data, 'text/css')
else:
root = parse(raw, line_numbers=False)
pretty_html_tree(None, root)
return etree.tostring(root, encoding=unicode)
def create_book(mi, path, fmt='epub', opf_name='metadata.opf', html_name='start.xhtml', toc_name='toc.ncx'):
''' Create an empty book in the specified format at the specified location. '''
path = os.path.abspath(path)
lang = 'und'
opf = metadata_to_opf(mi, as_string=False)
for l in opf.xpath('//*[local-name()="language"]'):
if l.text:
lang = l.text
break
lang = lang_as_iso639_1(lang) or lang
opfns = OPF_NAMESPACES['opf']
m = opf.makeelement('{%s}manifest' % opfns)
opf.insert(1, m)
i = m.makeelement('{%s}item' % opfns, href=html_name, id='start')
i.set('media-type', guess_type('a.xhtml'))
m.append(i)
i = m.makeelement('{%s}item' % opfns, href=toc_name, id='ncx')
i.set('media-type', guess_type(toc_name))
m.append(i)
s = opf.makeelement('{%s}spine' % opfns, toc="ncx")
opf.insert(2, s)
i = s.makeelement('{%s}itemref' % opfns, idref='start')
s.append(i)
CONTAINER = '''\
<?xml version="1.0"?>
<container version="1.0" xmlns="urn:oasis:names:tc:opendocument:xmlns:container">
<rootfiles>
<rootfile full-path="{0}" media-type="application/oebps-package+xml"/>
</rootfiles>
</container>
'''.format(prepare_string_for_xml(opf_name, True)).encode('utf-8')
HTML = P('templates/new_book.html', data=True).decode('utf-8').replace(
'_LANGUAGE_', prepare_string_for_xml(lang, True)
).replace(
'_TITLE_', prepare_string_for_xml(mi.title)
).replace(
'_AUTHORS_', prepare_string_for_xml(authors_to_string(mi.authors))
).encode('utf-8')
h = parse(HTML)
pretty_html_tree(None, h)
HTML = serialize(h, 'text/html')
ncx = etree.tostring(create_toc(mi, opf, html_name, lang), encoding='utf-8', xml_declaration=True, pretty_print=True)
pretty_xml_tree(opf)
opf = etree.tostring(opf, encoding='utf-8', xml_declaration=True, pretty_print=True)
if fmt == 'azw3':
with TemporaryDirectory('create-azw3') as tdir, CurrentDir(tdir):
for name, data in ((opf_name, opf), (html_name, HTML), (toc_name, ncx)):
with open(name, 'wb') as f:
f.write(data)
c = Container(os.path.dirname(os.path.abspath(opf_name)), opf_name, DevNull())
opf_to_azw3(opf_name, path, c)
else:
with ZipFile(path, 'w', compression=ZIP_STORED) as zf:
zf.writestr('mimetype', b'application/epub+zip', compression=ZIP_STORED)
zf.writestr('META-INF/', b'', 0755)
zf.writestr('META-INF/container.xml', CONTAINER)
zf.writestr(opf_name, opf)
zf.writestr(html_name, HTML)
zf.writestr(toc_name, ncx)
def beautify_text(raw, syntax):
from lxml import etree
from calibre.ebooks.oeb.polish.parsing import parse
from calibre.ebooks.oeb.polish.pretty import pretty_xml_tree, pretty_html_tree
from calibre.ebooks.chardet import strip_encoding_declarations
if syntax == 'xml':
root = etree.fromstring(strip_encoding_declarations(raw))
pretty_xml_tree(root)
elif syntax == 'css':
import logging
from calibre.ebooks.oeb.base import serialize, _css_logger
from calibre.ebooks.oeb.polish.utils import setup_cssutils_serialization
from cssutils import CSSParser, log
setup_cssutils_serialization(tprefs['editor_tab_stop_width'])
log.setLevel(logging.WARN)
log.raiseExceptions = False
parser = CSSParser(loglevel=logging.WARNING,
# We dont care about @import rules
fetcher=lambda x: (None, None), log=_css_logger)
data = parser.parseString(raw, href='<string>', validate=False)
return serialize(data, 'text/css')
else:
root = parse(raw, line_numbers=False)
pretty_html_tree(None, root)
return etree.tostring(root, encoding=unicode)
def commit_nav_toc(container,
toc,
lang=None,
landmarks=None,
previous_nav=None):
tocname, root = ensure_container_has_nav(container,
lang=lang,
previous_nav=previous_nav)
nav = ensure_single_nav_of_type(root, 'toc')
if toc.toc_title:
nav.append(nav.makeelement(XHTML('h1')))
nav[-1].text = toc.toc_title
rnode = nav.makeelement(XHTML('ol'))
nav.append(rnode)
to_href = partial(container.name_to_href, base=tocname)
spat = re.compile(r'\s+')
def process_node(xml_parent, toc_parent):
for child in toc_parent:
li = xml_parent.makeelement(XHTML('li'))
xml_parent.append(li)
title = child.title or ''
title = spat.sub(' ', title).strip()
a = li.makeelement(XHTML('a' if child.dest else 'span'))
a.text = title
li.append(a)
if child.dest:
href = to_href(child.dest)
if child.frag:
href += '#' + child.frag
a.set('href', href)
if len(child):
ol = li.makeelement(XHTML('ol'))
li.append(ol)
process_node(ol, child)
process_node(rnode, toc)
pretty_xml_tree(nav)
collapse_li(nav)
nav.tail = '\n'
if toc.page_list:
nav = ensure_single_nav_of_type(root, 'page-list')
nav.set('hidden', '')
ol = nav.makeelement(XHTML('ol'))
nav.append(ol)
for entry in toc.page_list:
if container.has_name(entry['dest']) and container.mime_map[
entry['dest']] in OEB_DOCS:
a = create_nav_li(container, ol, entry, tocname)
a.text = str(entry['pagenum'])
pretty_xml_tree(nav)
collapse_li(nav)
container.replace(tocname, root)
def add_stylesheet_links(container, name, text):
root = container.parse_xhtml(text, name)
head = root.xpath('//*[local-name() = "head"]')
if not head:
return
head = head[0]
sheets = tuple(container.manifest_items_of_type(lambda mt: mt in OEB_STYLES))
if not sheets:
return
for sname in sheets:
link = head.makeelement(XHTML('link'), type='text/css', rel='stylesheet', href=container.name_to_href(sname, name))
head.append(link)
pretty_xml_tree(head)
return serialize(root, 'text/html')
def add_stylesheet_links(container, name, text):
root = container.parse_xhtml(text, name)
head = root.xpath('//*[local-name() = "head"]')
if not head:
return
head = head[0]
sheets = tuple(container.manifest_items_of_type(lambda mt: mt in OEB_STYLES))
if not sheets:
return
for sname in sheets:
link = head.makeelement(XHTML('link'), type='text/css', rel='stylesheet', href=container.name_to_href(sname, name))
head.append(link)
pretty_xml_tree(head)
return serialize(root, 'text/html')
def set_landmarks(container, root, tocname, landmarks):
nav = ensure_single_nav_of_type(root, 'landmarks')
nav.set('hidden', '')
ol = nav.makeelement(XHTML('ol'))
nav.append(ol)
for entry in landmarks:
if entry['type'] and container.has_name(
entry['dest']) and container.mime_map[
entry['dest']] in OEB_DOCS:
a = create_nav_li(container, ol, entry, tocname)
a.set('{%s}type' % EPUB_NS, entry['type'])
a.text = entry['title'] or None
pretty_xml_tree(nav)
collapse_li(nav)
def pretty_print_opf(root):
from calibre.ebooks.oeb.polish.pretty import pretty_opf, pretty_xml_tree
pretty_opf(root)
pretty_xml_tree(root)
def create_book(mi,
path,
fmt='epub',
opf_name='metadata.opf',
html_name='start.xhtml',
toc_name='toc.ncx'):
''' Create an empty book in the specified format at the specified location. '''
if fmt not in valid_empty_formats:
raise ValueError('Cannot create empty book in the %s format' % fmt)
if fmt == 'txt':
with open(path, 'wb') as f:
if not mi.is_null('title'):
f.write(as_bytes(mi.title))
return
if fmt == 'docx':
from calibre.ebooks.conversion.plumber import Plumber
from calibre.ebooks.docx.writer.container import DOCX
from calibre.utils.logging import default_log
p = Plumber('a.docx', 'b.docx', default_log)
p.setup_options()
# Use the word default of one inch page margins
for x in 'left right top bottom'.split():
setattr(p.opts, 'margin_' + x, 72)
DOCX(p.opts, default_log).write(path, mi, create_empty_document=True)
return
path = os.path.abspath(path)
lang = 'und'
opf = metadata_to_opf(mi, as_string=False)
for l in opf.xpath('//*[local-name()="language"]'):
if l.text:
lang = l.text
break
lang = lang_as_iso639_1(lang) or lang
opfns = OPF_NAMESPACES['opf']
m = opf.makeelement('{%s}manifest' % opfns)
opf.insert(1, m)
i = m.makeelement('{%s}item' % opfns, href=html_name, id='start')
i.set('media-type', guess_type('a.xhtml'))
m.append(i)
i = m.makeelement('{%s}item' % opfns, href=toc_name, id='ncx')
i.set('media-type', guess_type(toc_name))
m.append(i)
s = opf.makeelement('{%s}spine' % opfns, toc="ncx")
opf.insert(2, s)
i = s.makeelement('{%s}itemref' % opfns, idref='start')
s.append(i)
CONTAINER = '''\
<?xml version="1.0"?>
<container version="1.0" xmlns="urn:oasis:names:tc:opendocument:xmlns:container">
<rootfiles>
<rootfile full-path="{0}" media-type="application/oebps-package+xml"/>
</rootfiles>
</container>
'''.format(prepare_string_for_xml(opf_name, True)).encode('utf-8')
HTML = P('templates/new_book.html', data=True).decode('utf-8').replace(
'_LANGUAGE_', prepare_string_for_xml(lang, True)).replace(
'_TITLE_', prepare_string_for_xml(mi.title)).replace(
'_AUTHORS_',
prepare_string_for_xml(authors_to_string(
mi.authors))).encode('utf-8')
h = parse(HTML)
pretty_html_tree(None, h)
HTML = serialize(h, 'text/html')
ncx = etree.tostring(create_toc(mi, opf, html_name, lang),
encoding='utf-8',
xml_declaration=True,
pretty_print=True)
pretty_xml_tree(opf)
opf = etree.tostring(opf,
encoding='utf-8',
xml_declaration=True,
pretty_print=True)
if fmt == 'azw3':
with TemporaryDirectory('create-azw3') as tdir, CurrentDir(tdir):
for name, data in ((opf_name, opf), (html_name, HTML), (toc_name,
ncx)):
with open(name, 'wb') as f:
f.write(data)
c = Container(os.path.dirname(os.path.abspath(opf_name)), opf_name,
DevNull())
opf_to_azw3(opf_name, path, c)
else:
with ZipFile(path, 'w', compression=ZIP_STORED) as zf:
zf.writestr('mimetype',
b'application/epub+zip',
compression=ZIP_STORED)
zf.writestr('META-INF/', b'', 0o755)
zf.writestr('META-INF/container.xml', CONTAINER)
zf.writestr(opf_name, opf)
zf.writestr(html_name, HTML)
zf.writestr(toc_name, ncx)
def create_book(mi, path, fmt='epub', opf_name='metadata.opf', html_name='start.xhtml', toc_name='toc.ncx'):
''' Create an empty book in the specified format at the specified location. '''
if fmt not in valid_empty_formats:
raise ValueError('Cannot create empty book in the %s format' % fmt)
if fmt == 'txt':
with open(path, 'wb') as f:
if not mi.is_null('title'):
f.write(mi.title)
return
if fmt == 'docx':
from calibre.ebooks.conversion.plumber import Plumber
from calibre.ebooks.docx.writer.container import DOCX
from calibre.utils.logging import default_log
p = Plumber('a.docx', 'b.docx', default_log)
p.setup_options()
# Use the word default of one inch page margins
for x in 'left right top bottom'.split():
setattr(p.opts, 'margin_' + x, 72)
DOCX(p.opts, default_log).write(path, mi, create_empty_document=True)
return
path = os.path.abspath(path)
lang = 'und'
opf = metadata_to_opf(mi, as_string=False)
for l in opf.xpath('//*[local-name()="language"]'):
if l.text:
lang = l.text
break
lang = lang_as_iso639_1(lang) or lang
opfns = OPF_NAMESPACES['opf']
m = opf.makeelement('{%s}manifest' % opfns)
opf.insert(1, m)
i = m.makeelement('{%s}item' % opfns, href=html_name, id='start')
i.set('media-type', guess_type('a.xhtml'))
m.append(i)
i = m.makeelement('{%s}item' % opfns, href=toc_name, id='ncx')
i.set('media-type', guess_type(toc_name))
m.append(i)
s = opf.makeelement('{%s}spine' % opfns, toc="ncx")
opf.insert(2, s)
i = s.makeelement('{%s}itemref' % opfns, idref='start')
s.append(i)
CONTAINER = '''\
<?xml version="1.0"?>
<container version="1.0" xmlns="urn:oasis:names:tc:opendocument:xmlns:container">
<rootfiles>
<rootfile full-path="{0}" media-type="application/oebps-package+xml"/>
</rootfiles>
</container>
'''.format(prepare_string_for_xml(opf_name, True)).encode('utf-8')
HTML = P('templates/new_book.html', data=True).decode('utf-8').replace(
'_LANGUAGE_', prepare_string_for_xml(lang, True)
).replace(
'_TITLE_', prepare_string_for_xml(mi.title)
).replace(
'_AUTHORS_', prepare_string_for_xml(authors_to_string(mi.authors))
).encode('utf-8')
h = parse(HTML)
pretty_html_tree(None, h)
HTML = serialize(h, 'text/html')
ncx = etree.tostring(create_toc(mi, opf, html_name, lang), encoding='utf-8', xml_declaration=True, pretty_print=True)
pretty_xml_tree(opf)
opf = etree.tostring(opf, encoding='utf-8', xml_declaration=True, pretty_print=True)
if fmt == 'azw3':
with TemporaryDirectory('create-azw3') as tdir, CurrentDir(tdir):
for name, data in ((opf_name, opf), (html_name, HTML), (toc_name, ncx)):
with open(name, 'wb') as f:
f.write(data)
c = Container(os.path.dirname(os.path.abspath(opf_name)), opf_name, DevNull())
opf_to_azw3(opf_name, path, c)
else:
with ZipFile(path, 'w', compression=ZIP_STORED) as zf:
zf.writestr('mimetype', b'application/epub+zip', compression=ZIP_STORED)
zf.writestr('META-INF/', b'', 0755)
zf.writestr('META-INF/container.xml', CONTAINER)
zf.writestr(opf_name, opf)
zf.writestr(html_name, HTML)
zf.writestr(toc_name, ncx)
def commit_nav_toc(container,
toc,
lang=None,
landmarks=None,
previous_nav=None):
from calibre.ebooks.oeb.polish.pretty import pretty_xml_tree
tocname = find_existing_nav_toc(container)
if previous_nav is not None:
nav_name = container.href_to_name(previous_nav[0])
if nav_name and container.exists(nav_name):
tocname = nav_name
container.apply_unique_properties(tocname, 'nav')
if tocname is None:
item = container.generate_item('nav.xhtml', id_prefix='nav')
item.set('properties', 'nav')
tocname = container.href_to_name(item.get('href'),
base=container.opf_name)
if previous_nav is not None:
root = previous_nav[1]
else:
root = container.parse_xhtml(
P('templates/new_nav.html', data=True).decode('utf-8'))
container.replace(tocname, root)
else:
root = container.parsed(tocname)
if lang:
lang = lang_as_iso639_1(lang) or lang
root.set('lang', lang)
root.set('{%s}lang' % XML_NS, lang)
nav = ensure_single_nav_of_type(root, 'toc')
if toc.toc_title:
nav.append(nav.makeelement(XHTML('h1')))
nav[-1].text = toc.toc_title
rnode = nav.makeelement(XHTML('ol'))
nav.append(rnode)
to_href = partial(container.name_to_href, base=tocname)
spat = re.compile(r'\s+')
def process_node(xml_parent, toc_parent):
for child in toc_parent:
li = xml_parent.makeelement(XHTML('li'))
xml_parent.append(li)
title = child.title or ''
title = spat.sub(' ', title).strip()
a = li.makeelement(XHTML('a' if child.dest else 'span'))
a.text = title
li.append(a)
if child.dest:
href = to_href(child.dest)
if child.frag:
href += '#' + child.frag
a.set('href', href)
if len(child):
ol = li.makeelement(XHTML('ol'))
li.append(ol)
process_node(ol, child)
process_node(rnode, toc)
pretty_xml_tree(nav)
def collapse_li(parent):
for li in parent.iterdescendants(XHTML('li')):
if len(li) == 1:
li.text = None
li[0].tail = None
collapse_li(nav)
nav.tail = '\n'
def create_li(ol, entry):
li = ol.makeelement(XHTML('li'))
ol.append(li)
a = li.makeelement(XHTML('a'))
li.append(a)
href = container.name_to_href(entry['dest'], tocname)
if entry['frag']:
href += '#' + entry['frag']
a.set('href', href)
return a
if landmarks is not None:
nav = ensure_single_nav_of_type(root, 'landmarks')
nav.set('hidden', '')
ol = nav.makeelement(XHTML('ol'))
nav.append(ol)
for entry in landmarks:
if entry['type'] and container.has_name(
entry['dest']) and container.mime_map[
entry['dest']] in OEB_DOCS:
a = create_li(ol, entry)
a.set('{%s}type' % EPUB_NS, entry['type'])
a.text = entry['title'] or None
pretty_xml_tree(nav)
collapse_li(nav)
if toc.page_list:
nav = ensure_single_nav_of_type(root, 'page-list')
nav.set('hidden', '')
ol = nav.makeelement(XHTML('ol'))
nav.append(ol)
for entry in toc.page_list:
if container.has_name(entry['dest']) and container.mime_map[
entry['dest']] in OEB_DOCS:
a = create_li(ol, entry)
a.text = str(entry['pagenum'])
pretty_xml_tree(nav)
collapse_li(nav)
container.replace(tocname, root)
def commit_nav_toc(container, toc, lang=None, landmarks=None, previous_nav=None):
from calibre.ebooks.oeb.polish.pretty import pretty_xml_tree
tocname = find_existing_nav_toc(container)
if previous_nav is not None:
nav_name = container.href_to_name(previous_nav[0])
if nav_name and container.exists(nav_name):
tocname = nav_name
container.apply_unique_properties(tocname, 'nav')
if tocname is None:
item = container.generate_item('nav.xhtml', id_prefix='nav')
item.set('properties', 'nav')
tocname = container.href_to_name(item.get('href'), base=container.opf_name)
if previous_nav is not None:
root = previous_nav[1]
else:
root = container.parse_xhtml(P('templates/new_nav.html', data=True).decode('utf-8'))
container.replace(tocname, root)
else:
root = container.parsed(tocname)
if lang:
lang = lang_as_iso639_1(lang) or lang
root.set('lang', lang)
root.set('{%s}lang' % XML_NS, lang)
nav = ensure_single_nav_of_type(root, 'toc')
if toc.toc_title:
nav.append(nav.makeelement(XHTML('h1')))
nav[-1].text = toc.toc_title
rnode = nav.makeelement(XHTML('ol'))
nav.append(rnode)
to_href = partial(container.name_to_href, base=tocname)
spat = re.compile(r'\s+')
def process_node(xml_parent, toc_parent):
for child in toc_parent:
li = xml_parent.makeelement(XHTML('li'))
xml_parent.append(li)
title = child.title or ''
title = spat.sub(' ', title).strip()
a = li.makeelement(XHTML('a' if child.dest else 'span'))
a.text = title
li.append(a)
if child.dest:
href = to_href(child.dest)
if child.frag:
href += '#'+child.frag
a.set('href', href)
if len(child):
ol = li.makeelement(XHTML('ol'))
li.append(ol)
process_node(ol, child)
process_node(rnode, toc)
pretty_xml_tree(nav)
def collapse_li(parent):
for li in parent.iterdescendants(XHTML('li')):
if len(li) == 1:
li.text = None
li[0].tail = None
collapse_li(nav)
nav.tail = '\n'
def create_li(ol, entry):
li = ol.makeelement(XHTML('li'))
ol.append(li)
a = li.makeelement(XHTML('a'))
li.append(a)
href = container.name_to_href(entry['dest'], tocname)
if entry['frag']:
href += '#' + entry['frag']
a.set('href', href)
return a
if landmarks is not None:
nav = ensure_single_nav_of_type(root, 'landmarks')
nav.set('hidden', '')
ol = nav.makeelement(XHTML('ol'))
nav.append(ol)
for entry in landmarks:
if entry['type'] and container.has_name(entry['dest']) and container.mime_map[entry['dest']] in OEB_DOCS:
a = create_li(ol, entry)
a.set('{%s}type' % EPUB_NS, entry['type'])
a.text = entry['title'] or None
pretty_xml_tree(nav)
collapse_li(nav)
if toc.page_list:
nav = ensure_single_nav_of_type(root, 'page-list')
nav.set('hidden', '')
ol = nav.makeelement(XHTML('ol'))
nav.append(ol)
for entry in toc.page_list:
if container.has_name(entry['dest']) and container.mime_map[entry['dest']] in OEB_DOCS:
a = create_li(ol, entry)
a.text = str(entry['pagenum'])
pretty_xml_tree(nav)
collapse_li(nav)
container.replace(tocname, root)
def pretty_print_opf(root):
from calibre.ebooks.oeb.polish.pretty import pretty_opf, pretty_xml_tree
pretty_opf(root)
pretty_xml_tree(root)